lattice scale argument !

Submitted by yaman on Wed, 10/19/2011 - 16:41

hallo ,
i am running simulation of a cube ,,
so i want to devide it with special number of lattices ,, but since that number is not big ,, i get so less particles ,
which let particles in flow domain ,,get through the wall boundary !! which is catastrophic mater !!!!:(
i tried to define the boundayr as a waal with wall/reflect ,, it didnt work although i satisfied the condition ( not periodic boundayr condition )
so what is the solution ,, how can i guarantee that my lattice resolution will create a real wall which will not let flow particles get through !!!

Forums:

ckloss's picture

ckloss | Wed, 10/19/2011 - 16:53

sorry, I have the feeling that my perception is insufficient to grasp enough information from your post to answer it appropriately. Could you please try to re-formulate your question?

Cheers, Christoph

yaman | Wed, 10/19/2011 - 18:14

the case is ,, in example :( liggghts_1p2p9/examples/LAMMPS/flow>in.flow.couette )
i changed the lattice scale from 0.7 to 0.01 , and i got in paraview : particles of flow ( middle domain , betwwen upper and lower walls ) are pentrating the walls !! and also the dimensions of the simulation box is changed during the simulation !!!
can i know why is this happening ?!
what is the suitable case for choosing lattice scale ?!
thankx